Comparative study between external standard method and internal standard method for low-level tritium measurements

Quenching is always present in water samples and the degree of quenching can vary from one sample to another even within the same batch. This means that quench correction should be carried out for each sample in order to determine the activity so that comparisons can be made between samples and other batches. In this paper, a comparative study of tritium measurements between two methods used to correct for quenching is presented. The methods used to determine counting efficiency in the presence of quenching are as follow: Spectral Quench Parameter of External standard method (SQP(E)) and Internal Standard Method. In this work, a low background liquid scintillation system detector (Quantulus 1220) is used to determine tritium activity concentration in heavy water with different concentration from 99,66 D/H+D% to 1,65 D/D+H %. A standard calibration curve for the SQP(E) technique has been carried out with 3H low level quenched PACKARD standard set that had an assayed value of 29,240 dpm/std + 1,6%. Quench correction for Internal Standard Method has been made for each sample of heavy water with Tritiated Water Internal Standard that had a tritium concentration of 2,51 x 106 dpm/g + 3,0%. A comparison between dilution factors, calculated both for D/(D+ H)% concentration and for tritium activity, measured by the two methods, is discussed in this paper. Internal Standard Method provides accurate results especially for lower D/(D+H)% concentration, which is the environmental samples case. Commercial standards set don't fulfill the exigency of an accurate environmental tritium measurement. It must consider the following problems: type of vial, type of scintillant, filled volume and counting geometry. Even if you make your own standard set for quenching calibration you must exceed another problem. The Compton electrons produced by external standard are energetic beta particles and do not have the same behavior of weak beta-particles in the sample itself. Hence a small amount quenching in tritium sample may remain undetected.(author)